Hi Jason,
Thanks, I'll look into this by more than skimming when I have more time.
It still does not really change points (1a), (1b), (2) and (3). An
advantage of participating in the ASF ought to be that these legal
matters should be discussed, and in this case discussed with Microsoft
by the ASF's legal counsel. It is my sincere hope and expectation that
the ASF has enough Intellectual Property savvy to resolve it. Once
resolved, then this resolution ought to be enough for the project to
vote +1 to release OOXML support in POI.
I truly believe that sourcesense is acting in good faith, and I also
believe that Andy is as well. Microsoft we all have good reason to
doubt.
I'm certainly -1 on releasing the OOXML branch until the issue is
resolved to the satisfaction of the ASF's legal counsel.
I'm +1 on work proceeding on the OOXML branch. If the next release
occurs without such assurance then I'm -1 on it until we are sure
nothing from OOXML has spilled over into the release.
This is important to resolve as significant progress has been made on
trunk. Enough so that a release there should happen soon.
But we aren't actually voting on this yet.
Regards,
Dave
On Apr 13, 2008, at 1:43 AM, Jason Harrop wrote:
Hi
On Sun, Apr 13, 2008 at 4:08 PM, David Fisher
<[EMAIL PROTECTED]> wrote:
As a preface, until I read the OSP when it was announced, I held a
similar
opinion to yours - afraid that Microsoft would squash POI, if given
the
opportunity. I'm not a lawyer, but I've read some legal documents
and had
some of my own IP protected. To me the OSP seems OK, but you have
asked us
to examine the issue.
I'm attaching Paul E. Merrell, J.D.'s view of the OSP.
cheers
Jason
---------- Forwarded message ----------
Date: Sun, Apr 13, 2008 at 4:04 PM
Subject: Re: Rejection of any ENCUMBERED Microsoft Donation to POI
To: Jason Harrop <[EMAIL PROTECTED]>
Hi, Jason,
The most authoritative source to cite for the Microsoft Open
Specification Promise not conferring sufficient rights to implement
OOXML is here. <http://law.bepress.com/unswwps/flrps/art71/>. If you
compare it with the earlier Groklaw EOOXML Objections analysis, you'll
see that the University of New South Wales law professors repeatedly
cite the EOOXML Objections document and agree with virtually every
point in it, but flesh it out with more legal research.
If you would like to read an absolutely comic response to the UNSW
report by a Microsoft lawyer who claims to have co-authored the OSP,
see <http://www.zdnetasia.com/news/software/
0,39044164,62037862,00.htm>.
Nothing but evasions, but ponder what it means to be a Microsoft
Regional Directory of Interoperability, especially if one is a lawyer
rather than an engineer. I was rocking in my chair laughing
uncontrollably when I read that job title. Yes, there are so many
regional aspects to interoperability, but methinks they have more to
do with IPR interop barriers than with the technology. :-) Are we
setting up regional Microsoft law offices to write and negotiate IPR
documents?
The portion of the EOOXML Objections document that is relevant is here
<http://www.groklaw.net/article.php?story=20070123071154671&query=eooxml+objections#Patent_rights_to_implement_the_Ecma_376_specification_have_not_been_granted
>.
It was published on January 23, 2007. I wrote the relevant sections
myself.
The significance of the earlier publication by Groklaw is that in all
the time since, Microsoft has made no response that addressed the
specific points raised. Their only apparent response to the EOOXML
Objections document was a short "legal memorandum" citing no legal
authorities that Microsoft solicited from the London office of a Swiss
law firm. <http://www.bakernet.com/NR/rdonlyres/CC54A6B6-79E8-4E0D-B290-C836D5F70867/0/OpenXML.pdf
>.
It addresses none of the specific criticisms, ignores the existence of
the criticism, and makes a series of grand but bald claims about how
great the OSP is. The memorandum does not meet Bar disciplinary
standards in the U.S. for memoranda giving legal advice to non-clients
and is in truth nothing but propaganda.
The OSP is merely a public relations tool. It confers no rights
whatsoever, but is deceptively worded to convey the impression to the
uneducated that it does. Microsoft's real licensing for OOXML is
RAND-Z negotiated, that is, royalty-free but with IPR restrictions
that have yet to be revealed and must be negotiated..Microsoft
acknowledgement of the RAND-Z licensing requirement for OOXML can be
found here. <http://openxmlcommunity.org/openxmlmyths.aspx#myth4>.
Look here for a short discussion by a Microsoft in-house standards
lawyer of what RAND-Z means. <http://standardslaw.com/wordpress/>.
Microsoft added to its level of deception by claiming that it had
submitted its OSP to ISO/IEC and that ISO/IEC had found the OSP
satsifactory. While it is true that Microsoft submitted a copy of the
OSP along with its IPR election form to ISO/IEC, it was a gratuitous
inclusion. ISO/IEC do not evaluate IPR, leaving that to the courts.
All that ISO/IEC require is that a particular form be filled out to
accompany disclosure of the patents relevant to a given draft
standard. In that form, the submittor promises, by checking boxes, to
make IPR available by one of three methods. The form is here.
<http://www.itu.int/dms_pub/itu-t/oth/04/04/T04040000020002PDFE.pdf>.
As stated in the form:
"However, while the Patent Holder in this situation is committing to
not charging any monetary amount, the Patent Holder is still entitled
to require that the implementer of the above document sign a license
agreement that contains other reasonable terms and conditions such as
those relating to governing law, field of use, reciprocity,
warranties, etc."
Beware the "etc." :-) Go here for a more detailed description of the
ISO/IEC Patent Policy.
<http://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/customview.html?func=ll&objId=3770791&objAction=browse
>.
In summary, ISO/IEC build a database of patents claimed to read on a
standard and require only the simple ambiguous promise. They do not
evaluate IPR. Moreover, Microsoft's selection of the "negotiated"
option implies that the OSP is not the real IPR document. I.e., if the
OSP were the real IPR document, there would be no need for
negotiation.
It bears notice that the RAND-Z form Microsoft submitted applies only
to Ecma 376 in the form it was submitted to JTC 1. A new one must be
submitted for the changed version adopted. I do not know whether that
has happened yet. My guess is not because the edited version is not
due out until around May 1. Likewise, at this point the Open
Specification Promise covers the Office 2003 XML Reference Schemas and
Ecma 376, but does not yet cover ISO/IEC:29500-2008.
<http://www.microsoft.com/interop/osp/default.mspx>. So as of tonight,
the OSP is irrelevant in any event.
I have been chasing iterations of the MS Office XML IPR documents
around the mulberry bush since 2005. See in addition to the EOOXML
Objections document linked above,
<http://www.groklaw.net/article.php?story=20050330133833843#A4>
(critiquing the Microsoft Office 2003 XML Reference Schemas patent
license) and <http://www.groklaw.net/article.php?story=20051129101457378#A2
>
(Covenant not to sue for the same schemas, later ineffectively
extended to cover OOXML as well).
In my entire legal career, I never encountered any legal documents as
weasel-worded as any of the three. They are simply reprehensible traps
for the unwary and absolutely unethical.
At the same time, I think Microsoft has irrevocably waived its rights
to assert its patents against OOXML implementations, but not because
of any of its IPR documents. Rather, I believe a court would find an
implied waiver in Microsoft's public statements about the OSP and the
degree of freedom it confers. The IPR documents themselves forbid
looking outside their four corners themselves as a source of rights,
so the public statements cannot be used to imply rights into the IPR
documents themselves.
However, this is an unavoidably risky opinion. Implied waivers do not
exist until a court implies them, so to speak.
In summary, I'll opine that any developer who plows hundreds or
thousands of hours into developing an implementation of OOXML without
a signed private agreement with Microsoft is building on legal
quicksand of unknown depth. I can't imagine that any of the big
software houses developing support for OOXML do not have separate
agreements with Microsoft allowing them to do so.
I stress that all of the above is personal opinion rather than legal
advice. I resigned from the Bar when I retired and am no longer
licensed to practice law.
Please feel free to share this post as widely as you wish.
Best regards,
Paul E. Merrell, J.D. ("Marbux")
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
